Maintaining Balance

Akron,OH 
Today
Sunny High 82FLow 62F
  • Temp: 77F
  • Hum: 61%
  • W: 4mph SW
SATURDAY
Moderate or heavy rain shower
High 83F
Low 60F
SUNDAY
Sunny
High 82F
Low 54F
MONDAY
Sunny
High 86F
Low 62F
TUESDAY
Patchy rain nearby
High 83F
Low 59F

Tuesday - May 21

Run
  • 20m
  • -----
  • -----

Just a quick workout. Time is a little short at the moment.

FORCED EVOLUTION

Another day in the trenches.  We're getting better.  Programmers and I only went to bed around 12/1 last night.

Was Axteraa right in thinking I was nuts in going live?  I still would disagree, even given the last, brutal 72 hours.

A little retrospective:  Up until now, BT had 10 years of bolt on-code.  The core always stayed the same, we never broke much when 'adding on' a new feature.  It was all carefully balanced, intertwined and it worked.  That code is 10 years old, there were many technical reasons to migrate some of our stuff to .NET, doing interopt stuff (not that I understand), etc.  To continue for another 10 years, we also had to update our core to make developing new features faster, etc

The majority of the bugs we are dealing with were fine on our dev2 server.  We were actually testing the training log for a month a least and worked out many, many things.  But as we found out, going to live 'just introduces things'.  And for technical reasons, we couldn't even test the Garmin functionality on the dev2 site.  We had to cross our fingers.

I knew that there were several, unoptimized items...BT blogger (that page was always a dog anyway), the nutrition log, possibly the friends page.  I also knew that by increasing the number of testers from 24 to 'several thousand' per day, we are going to find a lot more stuff.

A user indicated that 'we should have gotten another group of beta testers' and tested for more months.  I disagree.  It wouldn't have helped 'that much'.  They found the main bugs.  Their time was donated...there is only so much time people have.  I am grateful for the time they gave.  Prolonging it wouldn't have helped much more.

SURE, I knew several pages (nutrition, friends, etc) weren't 'ready for prime time' but that is JUST IT...what better way to REALLY improve upon a feature without a little FORCED EVOLUTION? Yeah, it sucks that it may be buggy for several days...but you will get a much better product then just me and a few eager beta testers trying to make it perfect.  What we get is a lot more ideas AND ESPECIALLY I get to 'really learn' how users use a particular item.  Take the Friends page. I have learned a great deal on how exactly that page is used by members, what they love about it.  I didn't have that information before.  Sure, I was very open in the redesign at the beginning, taking polls, showing screenshots, accepting comments (Whizz convinced me to put the dashboard back)...but there were only a few handfuls of people that took the time at that early stage to offer feedback.  I went on the information I had at the time.

So dumping on you a site that was, I would say, 95% complete (~80% optimized)...was a risk I took and I would take it again.  At the end of the day, the forums 'worked', the training log (aside from garmin) 'worked', the other pages 'worked' for the most part.  This was the time for the masses to find out all the other lurking bugs that n handful of beta testers with busy lives didn't find.  It was also a time for critical feedback on items (such as friends, nutrition log) so I could really understand how you use them and work on some quick changes so you have the best product once the bugs get worked out.

With ANY site that you took 10 years to add-on code, that you then have to overhaul for the future from the core to the user-interface, there will be A LOT of issues going live that you just won't catch in development.  You would need a team of about 6 programmers at least, running a completely new development live site not on a subdomain, on it's OWN server (not shared with the live site), with mirrored development environments on the servers and local machines...to possibly pull this off any better than we did.  That is A LOT of money and resources.  We are not a big company, just a couple of part-time programmers spread out over the country.  Anybody remember the TP upgrade a few years back?  I followed that, more for a learning experience.  Similar website, similar antiquated core and UI that we had.  They were a bigger company at the time, and pulling off their huge redesign had just as many issues, if not more, than what we have had recently.  The outcry was far-reaching as the interface changed even much more than ours.  MMR too.  One of their transformational redesigns left their equipment tracker on the old site and not on the live, beta site that users were defaulted too.  I knew better than to do that.  ;)

I know a portion of you think this was a terrible decision to go live with the bugs we have, but I politely disagree.  Given what we had to do, we pulled it off well and are quickly working out the 'oddities' and bugs that a few dozen eyes weren't ever going to catch, even on a geological time-scale.

Thank you much for your patience.  We are getting there.

  • Calories out: 2790 From lifestyle:2566, From activities:224

Monday - May 20

Swim
  • 17m 04s
  • 800.00 yards
  • 02m 08s /100 yards

4x200 on 5'. Only been managing one swim for the last couple of weeks.

Strength
  • 20m

  • EXERCISE

    Sets
    Min
    reps
    Max
    reps
    Min
    wt.(lb)
    Max
    wt.(lb)
  • GENERAL - Back lever - tuck 1
    3
    10
    10
    0
    0
  • GENERAL - Front Lever - tuck 1
    3
    10
    10
    0
    0
  • GENERAL - L-Sit Pullups (Rings)
    3
    13
    16
    170
    170
  • GENERAL - Planche - tuck 1
    3
    15
    15
    0
    0
  • CHEST - XR Pushup
    3
    15
    15
    0
    0
I have printed out the 'Friends' pages (screenshots) and taking them to my kids swim lessons to ponder.  There are sever easy solutions to put and inspire link under/around the profile pics.

  • Calories out: 2842 From lifestyle:2555, From activities:287

Sunday - May 19

I wasn't able to make it to my workout today.

Saturday - May 18

Come next week, all of this will be just a figment our your imagination.

(It may take a little longer)

Friday - May 17

Run
  • 25m
  • -----
  • -----

Neighborhood run while the site was down and processing a database script. I haven't done anything since Monday and needed to get something to log on the first day of the new site.

2:14 am at BT

60+ something confirmed bugfixes and enhancements today.  Damn, we they are good.  Nothing like a '...and this time I mean it!' deadline to get stuff out.

I think we are ready...ready as we will ever be.

I love my programmers.

  • Calories out: 2843 From lifestyle:2563, From activities:280
  • Health data: Hours slept: 7

Thursday - May 16

TODAY AT BT

Programmers meeting.  24 hours until we go live with the new site.  Axteraa just told me I was nuts.  Wink Perhaps.  I'm pretty confident in the state of things.  The programmers are fixing all outstanding items today and I will be verifying later.  A few things are hard to reproduce without getting more users online using it.  And with a site as complex as this, there is no way to really 'optimize it' for speed unless you...well...go live in the environment it will be operating in.  There are many things that I DO WANT MORE USER INPUT ON that I know we will have to make changes and further optimize.  I involved the users in many parts of the redesign - but that doesn't figure out all the issues. I will involve them in the site again once it's live and solicit feedback.  There will be several things that will become apparent that need changed.  

I FULLY REALIZE IT WON'T BE 'PERFECT'... but we can indeed spend another two months easily fine-tuning, optimizing, redoing, etc.  But there is a point where you just have to 'call it' as at the end of the day, no matter when we go live, there WILL be issues. Two programmers will be online all Friday and through the weekend, just working on any bugs that come up and improving speed.  

In essence, there WILL be issues Friday and through the weekend.  No matter how long we spend debugging with a subset of users on a dev site under no load, we will undoubtedly have issues going live and issues after we are live.  It's just that kind of intensive, data-driven site.  

Relax.  Don't worry.  We have great programmers who will make sure all is fine by early next week.

Wednesday - May 15

TODAY AT BT:

More additions to the bug lists.  While the kids were at their swim lessons, I noticed a few navigational issues that need corrected.

Ron's Training Log


 May 2013 
SunMonTueWedThuFriSat
    1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31 
 

Volume

Time
Distance

Actual vs Planned

Time

Upcoming races

Totals

  • July's totals
  • June's totals
  • 2014 totals
  • 2013 totals